Advertisement

How a New U.N. Advisory Group Wants to Transform AI Governance

A panel discussion at the World Governments Summit in Dubai on Feb. 13. From left: TIME Executive Editor Naina Bajekal, UAE Minister for AI Omar Sultan Al Olama, U.N. Secretary-General Technology Envoy Amandeep Gill, Spain's Former AI Minister Carme Artigas, and Kenyan Government Technology Adviser Philip Thigo Credit - Courtesy World Government Summit

Amid a smorgasbord of different national, multilateral and global attempts to regulate artificial intelligence (AI), it can be hard to take stock of all the rules that govern—or attempt to—the emerging technology.

The United Nations is attempting to rise above that morass, with its own effort to issue a set of recommendations for governance of AI. Its new advisory body, made up of 39 members from a wide range of countries and sectors, began working last fall to provide recommendations on the form and function of a new international agency for the governance of AI. The group has already issued an interim report on “Governing AI for Humanity,” which proposes a variety of measures to strengthen the governance of AI and recommendations on ethical AI deployment, accountability, and making AI more inclusive.

Read More: What the U.N.’s AI Advisory Group Will Do

The co-chair of the group in charge of drafting the U.N.’s AI rules warned on Tuesday that these kinds of recommendations must ultimately become binding lest they dissolve into insignificance. “We have the legitimacy of U.N. organizations…we are building on that trust; the U.N. charter, international law, and human rights,” said Carme Artigas, the co-chair of the U.N.’s high-level advisory body on AI and Spain’s former minister for digitalization and artificial intelligence. Emphasizing the importance of accountability, she added: “Whatever we do must be binding, otherwise it makes no sense.”

Artigas issued the warning during a panel discussion at the World Governments Summit in Dubai, which brought together several members of the U.N. advisory group to share their perspectives and was moderated by TIME executive editor Naina Bajekal. The body will publish its final report later this year, following an ongoing public consultation process, ahead of the U.N.’s Summit of the Future.

The UAE’s minister for artificial intelligence, digital economy and remote work applications, Omar Al Olama, welcomed the U.N.’s attempts to write a global set of recommendations that strike a balance between optimism and pessimism over the potential effects of the technology. “The U.N. has been a role model in mobilizing and generating consensus, in ensuring that regulation is fit for purpose, and in being pragmatic,” said Al Olama, who became the world’s first AI minister in 2017. “There are a lot of people that are fear-mongers, there are a lot of people that are naysayers, and there are a lot of people who are blinded by the opportunities and the optimism. But seeing that balance was refreshing.”

Members of the U.N. Advisory Body on AI discuss their recommendations for the governance of artificial intelligence at the World Governments Summit in Dubai on Feb. 13<span class="copyright">Courtesy World Governments Summit</span>
Members of the U.N. Advisory Body on AI discuss their recommendations for the governance of artificial intelligence at the World Governments Summit in Dubai on Feb. 13Courtesy World Governments Summit

For African nations that are not represented in other high-level international bodies seeking AI regulation, like the G7 and European Union, the U.N.’s process presents the most realistic chance at having a seat at the table for deciding on standards for the technology, said Philip Thigo, an AI adviser to the Kenyan government and a member of the U.N. advisory group. “A lot of countries in the Global South are not part of this global governance ecosystem,” Thigo said. “The United Nations remains the only multilateral global system that we have that creates opportunities for countries in the Global South.”

There is an urgent need, Thigo added, for AI development to account for linguistic and cultural differences to ensure that not only the Global North benefits from the technology. Underpinning current AI systems are vast quantities of data largely in the English language from North America, noted Amandeep Gill, the U.N. secretary general’s envoy on technology. That omits a large part of the world, he said. “It’s morally outrageous, and it’s also a practical issue, given the need for the diversity of data.”

Read More: How the U.N. Plans to Shape the Future of AI

Gill warned the room of government officials and business leaders from around the world that AI would not only reshape the world of tech, but also the practice of governance itself. And he argued for the importance of striking a balance between guarding against the downsides while protecting the space for innovation that could deliver tremendous benefits. “We don’t want to overregulate,” he said. “But if we don't manage the governance situation well, then we will lose the trust of the public.”

Write to Billy Perrigo at billy.perrigo@time.com.